Summary of Proposed Research: In an internally-controlled prospective study started in 1946-1947, an extensive body of data has been collected concerning the genetic, physiological, metabolic and psychological characteristics of 1337 Johns Hopkins medical students in classes graduating between 1948 and 1964. Now for the most part physicians, they have been under observation for 13 to 27 years. Follow-up information is recorded annually on subjects and parents, and death certificates are routinely obtained. Forty subjects have died between the ages of 21 and 57; 16 from suicide and 24 from other causes. The interval between entering Johns Hopkins and death ranged from a few weeks to 27 years; the median interval was 13 years. Initial data included detailed family history, extensive background data, measurements of blood pressure and heart rate at rest under stress (cold pressor test, exercise test, anoxemia test, smoking test), cholesterol and ponderal index. Habit Survey, Family Attitude questionnaire, Rorschach test and Figure Drawing test. Follow-up inquiries are directed toward the occurrance of disease and death in subjects and parents, with periodic checks on habits of diet, coffee, and alcohol, smoking and exercise, and career status. Objectives: 1. To identify the distinguishing characteristics in youth of the 16 future suicides and 24 other deceased subjects, using living classmates as controls. 2. To find new approaches to the prediction of suicide and other forms of premature death, with a view to prevention. Methods: The case study method will be used in conjunction with a variety of multivariate statistical techniques. Statistical methods include multiple discriminant function analysis, multiple regression analysis, canonical correlations, mini-max cluster techniques and image analysis.